Like Jev, Laya decides between fixed answer options, but it runs locally under Apache 2.0. I measured it on the same 200 banking queries as Jev, in English, in German, and with shuffled option order.

f451 combines the comfort of a wiki with Git as the single source of truth. Every page is a Markdown file on Forgejo or GitHub, every change goes through draft and review, and permissions come from th...

Jev pre-classifies tickets in the n8n workflow, and the local language model only decides when Jev is unsure. Getting there took labels of my own, a corrected priority matrix and a new prompt. Plus a ...

Jev from TypeSafe returns typed decisions with probabilities instead of text. I measured it on 200 real banking queries, 100 of them in direct comparison with four Claude models from Haiku 4.5 to Fabl...

Opus 5.5 always thinks, starts at effort medium, and has new quirks with long tasks, design, and safety filters. Anthropic's prompting guide, checked against the documentation and supplemented with my...

An introduction to agentic coding for beginners. We walk through the first hour with Claude Code on a small Python project, from the plan and the tests to CLAUDE.md, with real output, real costs, and ...

Anthropic promises that Opus 5.5 costs about 40 percent less than Opus 5. I ran both models at their default settings on ten verifiable tasks and measured tokens, time, and cost per solved task.

DeepReinforce trains a coding model that learns its own agent harness. I tested the 9B and 35B locally on an M3 Max. The idea is strong, the benchmarks are self-reported, and the biggest stumbling blo...

Agent memory is an architecture problem, not a toggle. An ordered overview: the cognitive memory functions, why the context window is not memory, RAG through GraphRAG with their real costs, and where ...

Gemma 4 12B is celebrated as the most complete local model. For Python I ran three variants against each other on an M3 Max, with real code execution, including a parser that defeats all three.

The series finale: n8n binds to its own service on SAP BTP with a real OAuth flow. Client Credentials against the xsuaa token endpoint, a CAP service as a SAP-native stand-in, the n8n OAuth2 credentia...

MTPLX promises noticeably faster local LLMs on the Mac via Qwen 3.6's built-in MTP heads. Measured on an M3 Max, with real numbers instead of marketing claims.

QuEra wants to ship a fault-tolerant quantum computer with more than 256 logical qubits in 2028. The lever is not exotic physics but error correction with unusually little overhead. An assessment with...

Microsoft celebrates Majorana 2 as a 1,000 times more stable quantum chip. Its foundation is Majorana quasiparticles, whose existence physics has not settled to this day. An assessment with technical ...

From: Europe (represented by Germany, Department of "We've Got an Idea, Actually" and on the ground by me David Krcek ) To: Anthropic, San Francisco Re: Unsolicited application for the role of locatio...

Operation goes production-ready. This article moves the single-process instance to queue mode: main accepts webhooks and enqueues executions in Redis, workers run them. With worker scaling, queue metr...

With @function you define your own CSS functions. The real novelty is not that CSS rebuilds Sass, but that a CSS function decides in the browser instead of at build time.

Three days after launch, an export-control directive forces Anthropic to disable Fable 5 and Mythos 5 for all customers. What happened, why, and what it means.

A simple 2-node n8n Error Workflow can alert you in Telegram, Slack, Google Chat, email, or another channel when business automations fail silently.

With the Containerization architecture and the new Container machine, Apple has spent two years building a native Docker alternative for the Mac. A technical analysis of who is really under attack.

ChatGPT sounds thoughtful. Claude sounds empathetic. And yet nobody is home. A mechanistic, philosophical and societal look at the difference between language models and consciousness.

Three days after launch, an export-control directive forces Anthropic to disable Fable 5 and Mythos 5 for all customers. What happened, why, and what it means.

The pipeline now enriches its tickets with master data from a SAP backend. This article shows the OData binding over HTTP, SAP's own error format, graceful degradation on an unknown customer, and wher...

The local agent runs without a key and without the cloud. But what does that independence rest on? Two releases from the same week, a cloud handoff from Anthropic and Apple's MLX stack, bring the ques...

MCP moves writing tools out of the agent into separate servers. We build a small MCP server, attach it to apfel, and measure what MCP solves and what it doesn't. Article 10 of the series "A Local Codi...

A single AI backend degrades to the rule-based emergency exit on failure. This article builds real redundancy from two local models: round-robin in normal operation, automatic failover on a single out...

The one-shot command becomes an interactive REPL session. The key to testability is the strict separation of render logic and terminal I/O. Article 9 of the series "A Local Coding Agent with apfel".

Two months after warning that Mythos was too dangerous to release, Anthropic makes the Mythos class public. A look at Claude Fable 5 and Mythos 5 between benchmark records, pricing tactics, and 30-day...

An overview of the WWDC26 Platforms State of the Union: the Foundation Models framework opens up to server models and goes open source, the new Core AI brings your own models on-device, and Xcode 27 p...

At WWDC26, Apple shows how MLX distributes large language models across multiple Macs. RDMA over Thunderbolt 5, the JACCL library, and a cluster of four M3 Ultras running a trillion-parameter model lo...

AI, Apple Silicon, Coding-Experimente, SwiftUI und Haltung.

Until now two separate paths ran side by side: the rule-based ingest and the AI classification. This article merges them into one pipeline, makes classification a swappable stage, and positions the ru...

The naive file edit fails on the small model. We build the agent around that weakness, with constrained output and a two-stage edit workflow, and measure how much can be recovered. Article 7 of the se...

A reproducible eval of 15 tasks across five categories measures where the local Foundation Model carries as a coding agent and where it breaks. With a cloud sample and a separate measurement of what t...

The AI classifier runs, but what happens when the model goes down? A global error workflow with a Telegram alert, an observability stack of Prometheus, Loki and Grafana, and a silent fallback that wan...

We build the first tools the agent uses to change things, and lock them down — a path sandbox, a confirmation gate, a diff before writing, and defense-in-depth for the shell. Article 5 of the series "...

Within a few weeks, two open-source tools have set out to solve Claude Code's memory problem. But they don't solve the same problem — Graphify builds a code graph, MemPalace stores the conversation.

We teach the agent to call tools — the tool definition in the OpenAI schema, the round-trip of invocation, execution and continuation, and a tool abstraction in Swift. Article 4 of the series "A Local...

Cross-document view transitions bring app-like navigation to classic multi-page sites — with a single CSS at-rule and no framework at all.